Project Description

Heart disease remains the number one cause of death worldwide, which takes an estimated 17.9 million lives every year and the annual direct and indirect costs are estimated at over $316.6 billion. Arrhythmia, also known as irregular heartbeat, is one of the very common heart diseases. In the U.S., more than 850,000 people are hospitalized for an arrhythmia every year. Common arrhythmia treatment includes medication and implantable devices (e.g., cardioverter defibrillator). However, the effect of these treatments varies among patients and can even induce lethal health risks if not used properly. For example, adverse effects of antiarrhythmic drugs not only can lead to life-threatening events but also are the main reason of drug withdrawal, causing over a billion economic loss associated with the cost of developing new drugs that end up with withdrawals. Thus, of paramount importance is predicting the potential adverse effects of drugs at early stages of development. However, studying the cause of cardiotoxicity with animal model-based preclinical trials remains challenging: these studies consider a small set of drugs that have low efficiency and accuracy. An alternative way to current practice is the use of cardiac models as a virtual platform to systematically study interactions between drugs and the heart. While advances have been made in the development of biophysically detailed cardiac models to understand drug-induced arrhythmias, the cause(s) of drug-induced cardiotoxicity and arrhythmia remain poorly understood due to the limitations of first-principle models. The objective of this undergraduate student research is to develop data-driven approaches to understand the interactions between antiarrhythmic drugs and heart function for 8 drug screening. This will bring machine learning tools, design of experiment and optimization into a unified framework to create a digital replicate of the heart for virtual depiction, which will provide students with a unique opportunity to learn knowledge and skills in engineering and computational cardiology. The research findings will be disseminated through journal and conference publications.